Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
BEAST was a 2011 chosen-plaintext attack that recovered session cookies from TLS 1.0 and SSL 3.0 connections by exploiting predictable initialisation vectors in CBC mode.

What Is the BEAST Attack?
BEAST — Browser Exploit Against SSL/TLS — is a chosen-plaintext attack published by Thai Duong and Juliano Rizzo at the ekoparty conference in September 2011. Tracked as CVE-2011-3389, it let an attacker who could observe a victim’s encrypted traffic and run script in the victim’s browser recover secrets such as session cookies, one byte at a time, from TLS 1.0 and SSL 3.0 connections.
The flaw was known in theory long before 2011. Researchers had described the problem with cipher block chaining initialisation vectors in TLS 1.0 as early as 2002, and TLS 1.1 fixed it in 2006. BEAST mattered because it turned a paper weakness into a working browser demonstration, at a time when the overwhelming majority of the web still ran TLS 1.0.
In cipher block chaining mode, each plaintext block is combined with the previous ciphertext block before encryption. The first block needs an initialisation vector instead, and TLS 1.0 reused the last ciphertext block of the previous record as the initialisation vector for the next one. Because that block travels over the wire in the clear, an attacker watching the connection knows the initialisation vector for the record that is about to be sent.
That turns the cipher into something an attacker can test guesses against. If you know the initialisation vector and you can control part of the plaintext, you can arrange for a block containing your guess to be encrypted and compare the result with a block you have already captured. A match confirms the guess.
Duong and Rizzo needed two things: a network position to capture ciphertext, and the ability to make the browser send chosen plaintext within the same TLS session as the secret. They obtained the second with a Java applet, later replicated with JavaScript and WebSockets, that issued requests to the target site with attacker-chosen path lengths. By shifting the boundary one character at a time, they lined up a block that contained exactly one unknown byte of the session cookie, then brute-forced that byte across at most 256 attempts.
Repeating the process walked through the cookie. The demonstration recovered a PayPal session cookie in a couple of minutes, which is what made the research land.
Against a current browser and a server offering TLS 1.2 or TLS 1.3, no. The attack is historical, and scanners that still report a BEAST finding are usually flagging the presence of TLS 1.0 rather than a genuinely exploitable path. It remains worth understanding for two reasons. First, it is the first in a family — CRIME and BREACH, Lucky 13, POODLE, and DROWN followed over the next five years — that collectively retired a generation of protocol versions. Second, legacy TLS 1.0 endpoints survive in places nobody looks: management interfaces, industrial controllers, payment terminals, and internal applications pinned to an old framework.
PCI DSS has required the disabling of early TLS since 30 June 2018, and both the NCSC and NIST advise TLS 1.2 as a minimum with TLS 1.3 preferred. Scan your estate for anything still negotiating TLS 1.0 or 1.1, confirm whether a dependency genuinely requires it, and plan the retirement rather than suppressing the finding. Pay particular attention to appliances and embedded devices, which rarely appear in application inventories but very often terminate TLS on an obsolete library.
Our infrastructure security testing services include protocol and cipher suite review across internal and external estates, so the forgotten TLS 1.0 listener is found and recorded.
In practice, no. Every maintained browser applies record splitting, and servers that matter negotiate TLS 1.2 or TLS 1.3. A vulnerability scanner reporting BEAST is almost always reporting that TLS 1.0 with CBC ciphers is still offered, not that an attack would succeed. The finding is still worth acting on, because TLS 1.0 brings other problems with it.
RC4 is a stream cipher, so it has no cipher block chaining and no initialisation vector to predict, which made it an easy short-term answer to BEAST in 2011. Subsequent research showed statistical biases in the RC4 keystream that allow plaintext recovery given enough ciphertext. RFC 7465 prohibited RC4 in TLS in February 2015, and it should not be enabled anywhere.
Two things at once: the ability to observe the victim’s encrypted traffic, for example on a shared or compromised network, and the ability to run code in the victim’s browser that makes requests to the target site. Needing both is why BEAST was always harder to use in the wild than its reputation suggested, and why browser-side mitigations were effective.
Test which protocol versions and cipher suites the endpoint offers, using a tool such as testssl.sh, sslscan, or an external TLS assessment service. If TLS 1.0 or 1.1 appears, or CBC cipher suites remain enabled on those versions, schedule removal. Include internal services, load balancers, mail gateways, and management interfaces, not only public web servers.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.